# Motor / Generator

In "Motor/Generator Analysis Redux" I wrote, but nobody replied to:
snipped-for-privacy@yahoo.com wrote:

Many thanks to Jim and other contributors to the thread "Motor/Generator Analysis".
I have put a lot of money and time into this, and I want to give it my best shot, but I don't want to whip a dead horse, so to say.
Frankly, I don't understand magnetics. At least not as I understand resonance. I'm an amateur musician; I understand resonance and know a little about phase shifts near the peak. I do understand that because the slope of the curve is negative on the high-frequency (capacitative)
But B x I makes my head spin. I'm fine in 3 dimensions. So I get some of it. And I get that in the cylindrical coordinate system, B and I can
be locally orthogonal, and can vary in time, with phase shifts, while being wrapped into a connected topology. I just don't feel that the way
I feel resonances. It's not intuitive.
Would replacing the rotor "windings" with copper wire or bus bar (easy), and rewinding the stator with bigger wire (hard) have any chance at all of working together by lowering the leakage inductance and rotor resistance to allow resonance?
That's my best question; is there any hope at all?
This is a one-off demo, not a production prototype!
Yours,
Doug Goncz Replikon Research Falls Church, VA 22044-0394
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On 22 Jun 2005 13:46:10 -0700, snipped-for-privacy@aol.com wrote:

This is a wound rotor motor?
Measure the DC resistance of the *stator* windings. If that is fairly low, then you have a chance. If it is a wound-rotor motor and the rotor is wound with fine wire, then I would think that replacing that with bussbars might significantly reduce impedance and slipspeed, and make it better-suited for what you are trying to do. If it's easy to do, I'd try it.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Wed, 22 Jun 2005 13:46:10 -0700, DGoncz wrote:

[some stuff]

Yes, I think there's hope, but first you must do some experiments and some study of basic electronics, I think.
If you are unable to interpret the results of your experiments so far, then all that that means is that you need to do some more study on basic electronics.
Google is your friend here. Try various keywords - find out what shows up!
Good Luck! Rich
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On 22 Jun 2005 13:46:10 -0700, snipped-for-privacy@aol.com wrote:

SNIP
You've been wading in pretty deep water - although, at first sight a self excited squirrel cage alternator sounds like a simple device they're a positive feedback system. They're so touchy that they're pretty well only used by dedicated amateurs - I don't know of a single commercial application.
Note - this refers only to SELF EXCITED alternators. The same machine driven at over synchronous speed and connected to a power grid works fine as an induction alternator and will feed back to the power grid the equivalent of it's mechanical input power less its generation losses. This works well in commercial wind power sytems.
I really don't think you'll have much luck unless you move to a decent size 4 pole machine and even then I doubt that you'd be very happy with the results. A small permanent magnet DC motor is soooo much easier and cheaper!
I'm sorry if this is a bit disappointing but it's all useful experience and, sadly, experience is usually gained the hard way. You picked a very tough one for yor first project - better luck next time.
Jim
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Fri, 24 Jun 2005 20:07:44 +0100, pentagrid wrote:

I just had a thought - I'm not that conversant with induction alternators, but I've heard the term "slip" - which, driving the shaft mechanically, faster than the synchronous speed, will generate power - now, my quesion is, if, say, you're on a windmill, the amount of slip will increase as the RPM increases, right? Is there some kind of formula or graph - I'd think that if the "slip" gets up to, like, 90 or 120 degrees, that your generator efficiency would go back down somewhere, or am I letting the drugs interfere with my common sense?
Thanks, Rich
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Speed= 0 Speed= Sync Speed= 2xSync Slip = 1 Slip = 0 Slip = -1 | | | | | +Torque | | | _ \|/ | | / \ | | | / \ | | |_____/ \ | | |____________\|_____________| | \ ______ | |\ / | | \ / | | \_/ -Torque /|\ /|\ /|\ |---Motoring--|--Generating-|
Plot a normal Torque-Speed curve from 0 to Sync Speed.
If the shaft is then driven faster than Sync then the Torque-Speed is a mirror image of the first quadrant.
--
Tony Williams.

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Sat, 25 Jun 2005 12:38:05 +0100, Tony Williams

That applies to grid-connected induction machines where grid frequency and number of poles in the machine determine the sync speed.
In self excited induction generators which are not grid connected then sync speed is not constant, it will vary with shaft speed.
Induction machines (motors or generators) can only operate with a lagging power factor (required to produce rotor field), typically about 0.8 at full load. The connected "load" must therefore have a leading power factor to match. In the case of grid connection generators and distributed power factor correction capacitors provide the required leading power factor. In the case of a self excited induction generator enough capacitance must be connected to insure the required leading power factor.
The required capacitance varies depending on shaft speed and electrical load. There are large commercial windmill generators out there which use proprietary controllers to optimize load power factor for maximum output power at any wind speed within range, rectify the output to a DC bus and invert to grid frequency. (There was an article on wind power in Mechanical Engineering magazine a few years ago which discussed these but I can't find it now).
While the mfgr's are not disclosing any details, I am fairly sure they are using a PFC type rectifier with the current control set point not exactly tracking voltage but leading it by an adjustable amount. Should be reasonable to adapt this strategy to a small battery charging windmill induction generator, at least for someone with a solid understanding of PFC methods. I would be inclined to start with a DSP implementation of a PFC controller like the one that TI provides sample code for as part of a UPS reference design, but other approaches are possible.
For the hobbiest who wants to experiment with induction generators without learning DSP PFC methods, simply connecting varying values of power factor correction capacitors should provide a means of adjusting output with changing speed, at least with a resistive load. Not sure how well this will work with a three phase diode rectifier battery charger, but it should work well if a PFC battery charger is used.
Sorry if the above is redundant, I missed most of this thread.
Glen
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Glen Walpert wrote:

I've been trying this without success. It's kind of like shooting a rifle in the dark at a squirrel whose position you don't know.
I'm planning on building a "universal" self-excited induction generator (SEIG) workstation. It will have an SPDT switch to put the motor to be tested across the line, or across a *monster* cap sub box in parallel with a GC Electronics 20-102 cap sub box I picked up on ebay for twenty bucks. GC's 20-102 design uses caps rated at 200 WVDC, enough for some AC experiments. If I were to hit resonance, and hold it by driving the motor at sync speed, they could blow, but (1) I probably won't hit resonance exactly, and (2) the motor will blow its load of flywheeling kinetic energy before it blows a cap (I hope). Caps do take a bit of surge.
There will be a place for an LCR meter, one I have or a better one in the future, and a test load, which can damp generator operation, as well as a ground, and leads to the 'scope and a voltmeter. I am not familiar enough with the literature to say this workstation is novel, but it may be. I certainly have never seen one in any lab. I intend to try the workstation first with my drill press / circular saw motor (remember the Crapmaster?), then Burden's motor.
My design load for the motor I am working with is *only* 8 watts. It's a white LED traffic signal used in rail and dockyard work, I believe, donated by John Viselli at Dialight (yes, John, I am still working on this) in exchange for a tip off on a likely stolen traffic light I bought on ebay, then had doubts about, and out of general high tech good will on John's part as well.
Now, this is a high tech load and should not damp generator operation, as it is rated 80-135 VAC unknown Hz, probably 50-60, maybe a wider range, and I expect, although I have not tried it with a Variac, that it will go off line while the generator output is building from about 1 VAC, which is what remanant magnetism will provide, through a few or dozens of cycles of resonance to some value above 80 VAC, at which point it should come on line and make use of the generated power.
And that is how the project looks today. A couple bits of perfboard 15x12 inches, a few supports about 9 inches long, and the whole thing will fit in an R-Kive box when I am not using it. The extended cap box will be 5-50 uf and 50-300 uf for 350 total, and the CG box is 11.111 uf for 361.1 uf or so.
Doug
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On 25 Jun 2005 08:40:05 -0700, snipped-for-privacy@aol.com wrote:

I trust you are using a three phase motor for your experiments - I have never heard of a sucessful single phase induction generator, and strongly suspect that it is either completely impossible or requires separate control of the start and/or phase-shifted run windings.
Suggest you study induction machine theory a bit before diving into cryogenic experiments and the like. "Principles of Alternating Current Machinery" by Ralph R. Lawrence, 4th Edition, 1953, is a classic on the subject and often available inexpensively from the usual used booksellers. This book discusses the theory of induction generators in detail, something rarely found in more recent texts.
Glen
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Thanks, Glen, that's an excellent reference, and I have ordered one Lawrence for \$18, book rate, to my PO Box.
I suggest
http://www.qsl.net/ns8o/Induction_Generator.html
for single phase induction generator practice, and I have no source for the theory.
Doug
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On 25 Jun 2005 17:54:38 -0700, snipped-for-privacy@aol.com wrote:

A bargain, I think you will like it. BTW, induction generators were discussed in the first edition of Lawrence, copyright 1916.

I took a look at the single phase induction generator info. Learn something new every day. The performance your source reports is fairly dismal however, starting a motor only 1/10 the nameplate rating, and providing significant output only very close to rated frequency and only into a well matched load, all no doubt due to the need to match the requirements of whatever phase shifting mechanism the motor uses to obtain a rotating field from single phase power.
The performance of 3-phase induction generators is sooooo much better, and used 3-phase motors are so cheap, that it only makes sense to use a single phase induction generator if you are going to run it from an engine with speed regulation, and connect only well matched loads. If you need to match widely varying speeds and powers such as a windmill generator, then 3-phase is clearly the only viable option, as it can operate well over a very wide range of speeds and loads.
Glen
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
snipped-for-privacy@yahoo.com wrote:

That's one I've been wondering about. I thought it should be as you say but had not come across anything to confirm it.
You really put out some good information concerning electricty- many thanks for it.
John

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Hey, gang.
snipped-for-privacy@yahoo.com wrote:

I've got a medium sized DC permag motor and not one but two inverters.
Is there a load-sharing or four quadrant inverter I could add to my collection to experiment with what I do already recognize as much simpler synchronous, non self-excited operation?
I'd drive the generator with the rear wheel as detailed at ftp://users.aol.com/DGoncz/Bicycle and in other posts, and invert the output as I did on January 1, 2005, to the hilarity of my neighbors, using the sine wave inverter this time, drive the motor on one phase, and collect power from the other, or, with a load-sharing or four quadrant inverter, put the ultracapacitors back on line between the generator and inverter, and use both phases of the motor as a synchronous generator.

Been there, done that. See ftp://users.aol.com/DGoncz for the video.
I want, in all this work, to *feel what it's like* in my muscles. I just have this incredible appetite to gage levels of power from 1 - 1000 watts as muscle power, in convoluted ways. I want to feel the positive feedback "kick in" when I pedal my generator. That's what I live for, and for the occasional stunt. It's like having an appetite for kinky sex, not that I would know what that is like.

No, it's OK.
If I replace the grease in my generator's bearings with light oil, it will rotate even more easily. This brings to mind the opportunity for winter operation. That brings to mind room temperature superconductors, which we don't have yet, but also the thermal coefficient of electrical resistivity, which is a known property. It may be at some low temperature, with the right bearing lube, R may indeed by less than sqrt(L*C), allowing resonant operation.
How low do I have to go?
That is, do I need a Dewar of Freon, Nitrogen, Helium, or Hydrogen to lower the coefficient of electrical resistivty to the point resonance is possible? Dr. Majewski would certainly support such an effort.
I'll go check now....
Doug
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Copy of letter to Dr. Majewski:
Dear Dr. Majewski:
I am going to send this email to Dr. Brandt at FSU.
Does it look OK?
Are you alright with the way I've stated our association?
There is a problem with the math. I get some very low sub-ohm figure for what R *should* be. Like 0.008 ohms. That doesn't hold out much hope against the possiblity of cryogenic generator operation, does it? I think not, but maybe I am doing something wrong.
Doug
Dear Dr. Brandt:
I have been developing a small, slow self-excited induction generator with Dr. Walerian Majewski at NVCC, nearby.
The problem I am up against now is that the stator DC resistance, R is still greater than the square root of the product of the winding inductance, L, and the resonant capacitor C. Units are ohms, henries, and farads. Values are 23 ohms, 136 millihenries, and 51 microfarads.
What I'd like to do is carefully dip a sample stator in whatever cryogens you have available in approximately five liter quantities, and measure the stator DC resistance at these boiling cryogen temperatures, to determine whether R might in fact be less than sqrt(L*C) at some temperature. If a "hit" is found, research can be conducted for a low temperature bearing lubricant, and a live test made, as the machine has only one moving part, plus two ball bearings.
Yours,
Doug Goncz Replikon Research Seven Corners, VA 22044-0394 Student member SAE snipped-for-privacy@aol.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Copy of letter to Dr. Brandt:
Dear Dr. Brandt:
I have been developing a small, slow self-excited induction generator (SEIG) with Dr. Walerian Majewski at NVCC, nearby.
The problem I am up against now is that the stator DC resistance, R is still greater than the square root of the product of the winding inductance, L, and the resonant capacitor C. Units are ohms, henries, and farads. Values are 23 ohms, 136 millihenries, and 51 microfarads. Capacitance to 361 microfards will be available soon.
What I'd like to do is carefully dip a sample stator in whatever cryogens you have available in approximately five liter quantities, and measure the stator DC resistance at these boiling cryogen temperatures, to determine whether R might in fact be less than sqrt(L*C) at some temperature. If a "hit" is found, research can be conducted for a low temperature bearing lubricant, and a live test made, as the machine has only one moving part, plus two ball bearings.
Frankly, I doubt this will happen, but room temperature superconductors are only an era away, and I could certainly present the resistance measurements at the SPS spring conference as an "experiment that failed, but produced results." :(
As far as I know, the literature contains temperature coefficents of resistance for small pieces of wire, not whole stators, although your lab has certainly got some big magnets, and I am sure you've characterized them thoroughly. There are solder joints in this stator, for example....
Yours,
Doug Goncz Replikon Research Seven Corners, VA 22044-0394 Student member SAE snipped-for-privacy@aol.com
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
snipped-for-privacy@aol.com wrote in

Here is a free idea for you:
Google on 'silver clay'. It fires to produce .999 fine silver.
Develop a way to extrude the wet silver clay, clad in a layer of insulating ceramic slip.
Extrude your conductors and form your windings while the materials are still flexable.
Make the end connections with silver clay slip. Fire the unit. [silver clay fires at a moderate red heat.]
You should have a low resistance set of windings.
--
bz

please pardon my infinite ignorance, the set-of-things-I-do-not-know is an
infinite set.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Dear Jim,
snipped-for-privacy@yahoo.com wrote:

A positive feedback system would be unstable.
I do understand that the initial remnant magnetization is fed back to produce more magentization, but since operation is on the capacitive side of the Bode plot, I see negative feedback as the primary characteristic.
That is, loading the generator incrementally produces reduced speed, producing more output. This is negative feedback, stable characteristic.
I would be very interested in hearing from you an explanation of which aspects of the system are positive-feedback in nature, and which are negative. Anyone else, also, is welcome to comment on this.
Doug
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On 28 Jun 2005 02:26:54 -0700, snipped-for-privacy@aol.com wrote:

It is a positive feedback sytem and it is unstable. With a sufficiently high loaded Q, feedback builds up the small residual magnetic pattern in the rotor to produce ever increasing output.
This suicidal tendency is only controlled by both shaft available input power (i.e. the speed drops) and the rising iron losses which drop the loaded Q and modify the feedback phase angle.
Jim